calibration-log
Skill0 starsChanged 12 days agoApache-2.0View on GitHub
Keep a dated, append-only record of research criteria and resolve them against primary evidence. Use when reviewing a thesis criterion, checking a past research statement, or recording a sourced resolution.
In this repository3 other files it gives its agents.
- opportunity-screenSkill
- source-checkSkill
- thesis-trackerSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Keep the log append-only. Corrections supersede an entry.
- Distinctive: line 15
Check whether the evidence available at registration supported the statement.
- Distinctive: line 23
Use explicit unknown states.
- Distinctive: line 27
Never output probabilities, price targets, position sizes or trading advice.
- Distinctive: line 28
Never backdate, delete a miss or reinterpret the criterion after resolution.
- Distinctive: line 29
Never register a forecast through a tool.
- Distinctive: line 30
Do not treat missing evidence as a met criterion.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Research resolution log
Record what was stated, when it was knowable, and what the primary source later showed. The existing skill name is retained for compatibility.
Workflow
- Record a precise research criterion, its registration time, resolution date and primary resolution source. Mark historical imports as retrospective.
- Resolve only after the source is available. Use met, not met, or cannot be checked. Preserve the original wording and cite the document and field.
- Keep the log append-only. Corrections supersede an entry; they never overwrite it.
- Check whether the evidence available at registration supported the statement. Keep later restatements distinct from the originally filed version.
- Report source gaps and unresolved criteria. Every displayed figure needs a linked, audited claim.
Output
| ID | Registered | Criterion | Resolves on | Primary source | Resolution | Evidence claim |
Use explicit unknown states. Keep advice and generated forecasts out of the log.
Rules
- Never output probabilities, price targets, position sizes or trading advice.
- Never backdate, delete a miss or reinterpret the criterion after resolution.
- Never register a forecast through a tool.
- Do not treat missing evidence as a met criterion.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

